Optimization of Process Parameters of Lagerstroemia Speciosa Seed Hull Pyrolysis Using a Combined Approach of Response Surface Methodology (RSM) and Artificial Neural Network (ANN) for Renewable Fuel Production

The present study addressed the valorization of Lagerstroemia speciosa seed hull (LS) biomass for the production of renewable fuel and chemicals via pyrolysis. Important pyrolysis parameters such as heating rate (H.R), temperature, and inert gas (N2) flow rate were optimized using the joint approach of Response Surface Methodology (RSM) and Artificial Neural Network (ANN). Results showed comparatively higher R2 and lower MSE value for ANN model than RSM. The experimental findings revealed that the optimum condition for the maximum bio-oil yield (45.6%) was: temperature = 550 degrees C, H.R = 65 degrees C/min, and N2 flow rate = 60 ml/min; however, at this condition, the predicted bio-oil yield using RSM and ANN was 44.98 and 45.10% respectively. The obtained bio-oil was characterized based on its physicochemical properties such as GCMS, FTIR, and 1H NMR. The current work provides an insight by combining both RSM and ANN modeling methodologies to get a more efficient way to process modeling.